1. Field of the Invention
This invention relates to electrochemical cells designed to synthesize peroxide (HOO.sup..theta., O.sub.2.sup..theta., or H.sub.2 O.sub.2 dissolved in an alkaline medium) by cathodic reduction of an oxygen-containing gas and to processes for operating such cells. An aspect of this invention relates to electrochemical cells for the synthesis of peroxide wherein the electrolyte is divided into an aqueous alkaline catholyte and an aqueous alkaline anolyte. Another aspect of this invention relates to a process for synthesizing peroxide which can be operated at relatively low cell voltages and relatively high current densities and efficiencies.
2. Description of the Prior Art
It has long been known that hydrogen peroxide can be synthesized electrochemically, taking advantage of modern advances in electrochemical cell technology. The patent literature published on this subject in the late 1960's and early 1970's took into consideration the possibility of using a gas-diffusion cathode. A "gas-diffusion electrode" is normally considered to comprise a structure which is gas permeable on one major surface (sometimes called the "gas side") and electrocatalytic on the opposite major surface, which opposite surface is in contact with the electrolyte and is sometimes called the "electrolyte side". The electrolyte is permitted to permeate into the electrolyte side to a degree sufficient to provide a multi-phase interface between a gaseous reactant, a solid electrocatalytic material, and the electrolyte (which is generally a liquid). However, significant permeation of electrolyte through pores or interstices within the catalytic material of the gas-diffusion electrode at a significant flow rate is neither necessary nor desirable.
For a representative sampling of disclosures from this late-1960's/early 1970's period, see the several U.S. patents issued to Grangaard, e.g. U.S. Pat. No. 3,459,652 Aug. 5, 1969), U.S. Pat. No. 3,462,351 Aug. 19, 1969), U.S. Pat. No. 3,507,769 (Apr. 21, 1970), U.S. Pat. No. 3,592,749 (Jul. 13, 1971), and U.S. Pat. No. 3,607,687 (Sep. 21, 1971), These disclosures typically contemplate a generally free-flowing catholyte which takes up peroxide (generally in the form of HO.sub.2.sup..theta. dissolved in the alkaline catholyte) and is withdrawn from the cell for the purpose of recovering a product which is intended to be directly useful in industry, e.g. as an alkaline bleach solution.
Experts in the electrochemical synthesis art found the performance of the Grangaard cells to be very disappointing, however, and by the mid-1970's, even the fundamental principles upon which the Grangaard concepts were based were being called into question. For example, according to Oloman and his coworkers, see U.S. Pat. Nos. 3,969,201 and 4,118,305, issued Jul. 13, 1976 and Oct. 3, 1978, respectively, the Grangaard cells produced an aqueous alkaline product having a peroxide concentration of only about 0.5% with an NaOH/H.sub.2 O.sub.2 ratio (by weight percent) of 4:1 (of U.S. Pat. No. 3,459,652). As is known in the art, some uses of bleaching solution, e.g. in the pulp and paper industry, generally call for much higher concentrations of peroxide and/or for NaOH/H.sub.2 O.sub.2 ratios in the range of about 1:1 to about 2:1. Oloman et al, among others, questioned the basic idea of utilizing a gas-diffusion cathode of the classical structure wherein catholyte merely permeates into the electrode structure from the electrolyte side. Thus, by the mid- to late 1970's, prior art workers were directing their attention to cathode structures constructed from a fluid-permeable, electrically conductive mass (e.g. a bed of conductive catalytic particles or a fixed, porous conductive catalytic matrix) with sufficient porosity to permit a constant trickle or flow of electrolyte through the entire volume (or most of the volume) of the cathode mass. In an electrochemical cell provided with such a fluid-permeable, electrically conductive cathode mass, the cathode can, if desired, fill up the entire cathode compartment, so that all or most of the catholyte is confined to the interior of the cathode mass.
In subsequent developments based upon the packed-bed or porous matrix concept of a cathode, the cathode was in some cases placed in contact with a non-conducting porous matrix (such as a felt) or was employed in a cell having in essence a single electrolyte rather than an electrolyte divided into catholyte and anolyte.
In many patent disclosures illustrating the packed-bed or porous matrix concept of a cathode, the product (generally an alkaline solution of peroxide, most typically catholyte which has been passed through the cathode mass) is collected from an end or edge or other portion of the cathode mass, rather than from a generally free-flowing catholyte which has merely contacted and/or permeated to some degree a surface of the cathode. Alternatively, the product is essentially catholyte which has wicked through a non-conductive, porous mass such as a felt which is in contact with the cathode.
Peroxide-generating cells containing packed-bed or porous-matrix cathodes have in recent years become commercially available for use as on-site peroxide generators, and considerable effort has gone into the optimization of their performance. However, these commercially available cells operate at overall cell potentials (E.sub.cell) of about 2.0 V and current densities not significantly exceeding 60 amperes per square foot (about 64.5 mA/cm.sup.2 =645 A/m.sup.2). Even assuming a current efficiency of 85 to 90%, a large amount of electrode surface area is required in a typical commercial installation, resulting in higher capital costs to the industrial user.
Moreover, the packed-bed or porous matrix concept of cathode construction has not provided any improvement in quality control as compared to cells utilizing gas-diffusion cathodes. The packed bed or porous matrix can develop "hot spots" in which current densities, etc. are higher than average for the bed or matrix as a whole, thereby creating the risk that some part of the cathode might become "starved" for three-phase interface sites and can place the entire bed or matrix at risk of catastrophic failure. This risk can be reduced through the use of a significant stoichiometric excess of oxygen, but non-uniform consumption of oxygen throughout the cathode bed or matrix still contributes to poor quality control. In addition, bipolar cell construction, with stacking of cells for more efficient overall operation, is problematic, due to the variability in the performance of individual cells.
Accordingly, despite significant advances in the field of on-site electrosynthesis of peroxide over the last twenty years, cell performance is still in need of substantial improvement.